Contact: moussa [dot] 45 [at] osu [dot] edu
Google Scholar | LinkedIn | X | GitHub
Contact: moussa [dot] 45 [at] osu [dot] edu
Google Scholar | LinkedIn | X | GitHub
I'm Hanane Nour Moussa, a third year computer science Ph.D. student at the Ohio State University advised by Dr. Huan Sun and Dr. Sachin Kumar -- hence part of the Sun ☀️ and SKAI 🪂 labs!
My research interests broadly cover AI for code, reasoning, and scientific applications. In particular, my current research goals are:
Building data and infrastructure to advance AI capabilities for scientific coding.
Advancing core AI capabilities relevant to scientific applications, including memory, tool use, and long-context understanding.
Evaluating AI systems on real-world scientific workflows.
Studying and addressing safety risks while improving human–AI interaction and trust in scientific settings.
Feel free to contact me for collaborations or for questions/chat about my work!
News
[October 2026] I am co-organizing LM4Sci 2.0 Workshop and moderating a panel at COLM 2026 in SF, and presenting D3-Gym at EMNLP 2026 in Budapest, Hungary.
[August 2026] Third year of PhD starts with D3-Gym accepted to EMNLP 2026 (Main) and nominated for an award!
[May 2026] I'll be spending Summer 2026 interning with Oracle Machine Learning Research Group in Burlington, MA working with Jacob Beck and Ari Kobren.
[April 2026] D3-Gym is released! Paper, code & data.
[November 2025] Our new work ScholarEval on literature grounded research idea evaluation is released!
[October 2025] AutoSDT won best paper award at LM4Sci@COLM!
[October 2025] Headed to COLM 2025 to present AutoSDT at LM4Sci workshop.
[August 2025] Second year of Ph.D. starts!
[August 2025] AutoSDT accepted to EMNLP 2025 Main!
[August 2025] Gave a talk at Ai2 Aristo about automated research idea evaluation. Was great to interact and have thought-provoking conversations with fellow researchers interested in AI4Science!
[July 2025] AutoSDT accepted to COLM LM4Sci workshop!
[July 2025] Mind2Web 2 released! Happy to have joined this huge effort to rigorously evaluate web agents on realistic, long horizon tasks. Paper here.
[June 2025] Newly released paper! AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists. arXiv, data, code, website.
[August 2024] Started my Ph.D. at OSU!